Filter out more invalid URIs that make PyLD crash

swh/indexer/metadata_dictionary/base.py

@@ -6,6 +6,7 @@
 import json
 import logging
 from typing import Any, Callable, Dict, List, Optional, Tuple, TypeVar
+import urllib.parse
 import uuid
 import xml.parsers.expat
 
@@ -240,11 +241,18 @@ class DictMapping(BaseMapping):
                     for item in v:
                         graph.add((root, codemeta_key, rdflib.Literal(item)))
                 elif k in self.uri_fields and isinstance(v, str):
-                    graph.add((root, codemeta_key, rdflib.URIRef(v)))
+                    # Workaround for https://github.com/digitalbazaar/pyld/issues/91 : drop
+                    # URLs that are blatantly invalid early, so PyLD does not crash.
+                    parsed_url = urllib.parse.urlparse(v)
+                    if parsed_url.netloc:
+                        graph.add((root, codemeta_key, rdflib.URIRef(v)))
                 elif k in self.uri_fields and isinstance(v, list):
                     for item in v:
                         if isinstance(item, str):
-                            graph.add((root, codemeta_key, rdflib.URIRef(item)))
+                            # ditto
+                            parsed_url = urllib.parse.urlparse(item)
+                            if parsed_url.netloc:
+                                graph.add((root, codemeta_key, rdflib.URIRef(item)))
                 else:
                     continue
 

swh/indexer/tests/metadata_dictionary/test_npm.py

@@ -294,7 +294,7 @@ def test_npm_repository_normalization():
     }
 
 
-def test_npm_empty_uris():
+def test_npm_invalid_uris():
     package_json = rb"""{
   "version": "1.0.0",
   "homepage": "",
@@ -344,6 +344,22 @@ def test_npm_empty_uris():
         "version": "1.0.0",
     }
 
+    package_json = rb"""{
+  "version": "1.0.0",
+  "homepage": "http:example.org",
+  "author": {
+    "name": "foo",
+    "url": "http:example.com"
+  }
+}"""
+    result = MAPPINGS["NpmMapping"]().translate(package_json)
+    assert result == {
+        "@context": "https://doi.org/10.5063/schema/codemeta-2.0",
+        "type": "SoftwareSourceCode",
+        "author": [{"name": "foo", "type": "Person"}],
+        "version": "1.0.0",
+    }
+
 
 @settings(suppress_health_check=[HealthCheck.too_slow])
 @given(json_document_strategy(keys=list(MAPPINGS["NpmMapping"].mapping)))  # type: ignore